Duration of Venous Thromboembolism Risk Across a Continuum ... · ORIGINAL RESEARCH Duration of Venous Thromboembolism Risk Across a Continuum in Medically Ill Hospitalized Patients

ORIGINAL RESEARCH

Duration of Venous Thromboembolism Risk Across a Continuum inMedically Ill Hospitalized Patients

Alpesh N. Amin, MD, MBA, FACP1*, Helen Varker, BS2, Nicole Princic, MS2, Jay Lin, PhD3, Stephen Thompson, MS4,

Stephen Johnston, MA2

1School of Medicine, University of California-Irvine, Irvine, California; 2Thomson Reuters, Washington, District of Columbia; 3Novosys Health,Flemington, New Jersey; 4sanofi-aventis, Bridgewater, New Jersey.

BACKGROUND: Patients hospitalized for medical illnessare at increased risk of venous thromboembolism (VTE), butthe duration of risk is not well understood.

OBJECTIVE: To assess incidence and time course ofsymptomatic VTE following hospitalization for medicalillness in a large, real-world patient population.

DESIGN: Data were extracted from the Thomson ReutersMarketScanV

R

Inpatient Drug Link File.

PATIENTS: Those hospitalized with cancer, heart failure,severe lung disease, or infectious disease from 2005 to 2008.

MEASUREMENTS: The cumulative VTE risk over 180 daysafter admission was calculated using Kaplan-Meieranalysis. VTE hazard was calculated on a daily basis andsmoothed through LOESS regression.

RESULTS: The analysis included 11,139 medically ill patients,46.7% and 8.8% of whom received pharmacologicalthromboprophylaxis during hospitalization and after discharge,

respectively. The mean duration of prophylaxis duringhospitalization was 5.0 days. Of the 11,139 patients, 366(3.3%) experienced a symptomatic VTE event. VTE eventswere most frequent during days 0-9 (97 events), followed bydays 10-19 (82 events). The mean length of hospital stay was5.3 days, and 56.6% of all VTE events occurred afterdischarge. VTE hazard peaked at day 8, with 1.05 events per1000 person-days.

CONCLUSIONS: The time course of VTE in medical patientsshows that risk of symptomatic VTE is highest during the first19 days after hospital admission, and extends into the periodafter discharge. Future research is warranted to investigaterisks and benefits of reducing the incidence of VTE afterdischarge, including the role of improving thromboprophylaxispractices in the inpatient setting and extendingthromboprophylaxis after hospitalization. Journal of HospitalMedicine 2012;7:231–238.VC 2011 Society of Hospital Medicine

Patients who are hospitalized for acute medical illnessare at an increased risk of developing venous throm-boembolism (VTE), which comprises deep-vein throm-bosis (DVT) and pulmonary embolism (PE).1–3 In arecent real-world study of 158,325 US medical patientsby Spyropoulos et al,4 4.0% of patients developedDVT, 1.5% developed PE, and 0.2% developed bothDVT and PE. Furthermore, results from a population-based case-control study indicate that hospitalizationfor medical illness accounted for a proportion of VTEevents similar to that of hospitalization for surgery(22% and 24%, respectively).5

Thromboprophylaxis reduces VTE incidence in at-risk medical patients and is recommended accordingto evidence-based guidelines from the American Col-lege of Chest Physicians (ACCP).1 The ACCP guide-

lines advocate that acutely ill medical patients admit-ted to the hospital with congestive heart failure (CHF)or severe lung disease/chronic obstructive pulmonarydisease (COPD) or those who are confined to bed andhave one or more additional risk factors (includingactive cancer, previous VTE, sepsis, acute neurologicdisease, or inflammatory bowel disease) receive phar-macological prophylaxis with low–molecular weightheparin (LMWH), low-dose unfractionated heparin(UFH), or fondaparinux.1 Although guidelines providerecommendations for the duration of prophylaxis aftermajor orthopedic surgery, such recommendations areunavailable for medical patients. In clinical trials ofacutely ill medical patients, prophylaxis regimensfound to be effective were provided for a duration ofhospitalization of 6-14 days.6–8 The mean length ofhospital stay for medical illnesses is decreasing and iscurrently shorter than 6-14 days.9,10

In clinical practice, the duration of VTE risk duringand after hospitalization is not well understood inmedical patients, particularly in the context of short-ening hospital stays. Such information could, how-ever, provide insight into whether current thrombo-prophylaxis practices reflect real-world need. To gaina greater understanding of the period during whichpatients are at risk of VTE, this retrospective,

*Address for correspondence and reprint requests: Alpesh N. Amin,MD, MBA, FACP, University of California-Irvine, 101 The City Drive South,Building 58, Room 110, ZC-4076H, Orange, CA 92868; Tel.:714-456-3785; E-mail: [email protected]

Additional Supporting Information may be found in the online version ofthis article.

Received: March 30, 2011; Revised: September 30, 2011; Accepted:October 28, 20112011 Society of Hospital Medicine DOI 10.1002/jhm.1002Published online in Wiley Online Library (Wileyonlinelibrary.com).

An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 7 | No 3 | March 2012 231

observational study assessed the incidence and timecourse of symptomatic VTE events during and afterhospitalization in a large population of US medicalpatients.

METHODSData and Patient Selection

This study employed linked administrative claims dataand hospital billing data contained in the ThomsonReuters MarketScanV

R

Inpatient Drug Link File. Thiscombines longitudinal patient-level inpatient and out-patient medical and pharmaceutical claims data fromthe MarketScan Commercial claims data from theMarketScan Commercial Claims and Encounters(Commercial) and Medicare Supplemental and Coor-dination of Benefits (Medicare Supplemental) data-bases, with hospital discharge records detailing serv-ices used and drugs administered during ahospitalization, which are included in the HospitalDrug Database. The linked data sources enable analy-sis of a patient’s experience before, during, and after ahospitalization. The present study was not designed toobtain bleeding rates.The study cohort comprised patients considered to

be at-risk for VTE as a result of a medical hospitaliza-tion occurring between January 1, 2005, and Decem-ber 31, 2008. At-risk medical hospitalizations werethose for which the primary diagnosis was for cancer,CHF, severe lung disease/COPD, or infectious disease(see Supporting Information, Appendix I, for Interna-tional Classification of Diseases, 9th Revision, ClinicalModification [ICD-9-CM] codes used to identifypatients with medical illnesses). Included patientswere required to be at least 18 years of age at thetime of admission and were required to be continu-ously enrolled in their insurance benefits for at least12 months before admission (the baseline period) andfor at least 180 days after the admission date (theevaluation period) to ensure that all administrativeclaims data during that period were captured. Patientswho died in-hospital from any cause were exemptedfrom the continuous enrollment criterion, as long asthey had been continuously enrolled prior to inpatientdeath. Patients transferred from or discharged toanother acute-care facility were excluded because ofthe possibility for incomplete inpatient data capture.For patients who had multiple medical hospitaliza-tions between January 1, 2005, and December 31,2008, the hospitalization around which the analysisfocused was randomly selected from the set of poten-tial medical hospitalizations.

Prophylaxis

Pharmacological prophylaxis was identified via chargecodes during hospitalization or via pharmacy claimsafter discharge for UFH, enoxaparin, dalteparin, war-farin, and fondaparinux. All dosages of a pharmaco-logical agent were considered prophylactic only if

there was no evidence of VTE during the admission,with the exception of warfarin (Supporting Informa-tion, Appendix II). Post-discharge use of anticoagula-tion therapy was measured for up to 35 days after dis-charge from the hospital. Use of mechanicalprophylaxis during hospitalization was identified viacharge codes for graduated compression stockings andcharge codes indicating use of intermittent pneumaticcompression devices and/or venous foot pumps. Theappropriateness of prophylaxis was not assessed.

AnalysisThe risk of VTE was estimated across an evaluationperiod of 180 days by measuring VTE occurrence andperson-time exposure. Inpatient VTE occurrence wasdefined as any nonprimary diagnosis of DVT and/orPE during the at-risk hospitalization. VTE after dis-charge was defined as an ICD-9-CM diagnosis code,whether primary or secondary, for DVT or PE in theevaluation period during an emergency room or inpa-tient admission, or on an outpatient claim with 1 ormore of the following confirmatory events: an emer-gency room or inpatient admission for VTE within 2days of the outpatient diagnosis; a prescription claimfor enoxaparin, fondaparinux, or UFH within 15 daysafter diagnosis; or a prescription claim for warfarinwithin 15 days after diagnosis and no evidence ofatrial fibrillation or atrial flutter in the 6 monthsbefore the outpatient diagnosis for DVT or PE. Per-son-time exposure was measured as the number ofdays from the hospital admission date to the firstoccurrence of VTE, or censoring at a subsequent at-risk hospitalization, death, or 180 days afteradmission.Cumulative risk of VTE over the 180-day evaluation

period was calculated by the Kaplan-Meier productlimit method of survival analysis and displayed fordeciles of cumulative risk at 180 days after the hospi-tal admission date. The risk of VTE at each point oftime during the evaluation period (the hazard func-tion) was first calculated on a daily basis and thensmoothed via LOESS regression, a locally weightedregression procedure.

RESULTSPatient Demographics

A total of 11,139 medical patients were included inthe analysis (Figure 1), with a mean 6 standard devia-tion (SD) age of 67.6 6 13.9 years, and 51.6% werewomen (Table 1). Of the reasons for admission to thehospital, 51.5% of patients were admitted for severelung disease/COPD, 20.1% were admitted for cancer,15.3% were admitted for CHF, and 13.1% wereadmitted for severe infectious disease. Most patientswere treated in an urban hospital (87.5%), in a hospi-tal without teaching status (87.9%), and in the SouthCensus region (74.1%). The majority of patients weretreated in medium-sized to large care facilities. Risk

Amin et al. | Duration of VTE Risk in Medically Ill Patients

232 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 7 | No 3 | March 2012

factors for VTE during the baseline period includedhospitalization for a medical condition with a highrisk for VTE (75.6%), a prior at-risk hospitalization(18.6%), cancer therapy (10.0% of all medicalpatients combined and 18.5% of cancer patients),trauma (9.2%), and previous VTE (4.3%).

VTE Prophylaxis

Patients stayed in hospital for a mean 6 SD durationof 5.3 6 5.3 days, varying from 4.6 6 3.9 days inpatients with CHF to 6.7 6 6.5 days in patients withinfectious disease, during which 46.7% of patientsreceived pharmacological VTE prophylaxis. Inpatientpharmacological prophylaxis rates ranged from64.1% in patients with CHF to 30.7% in patientswith cancer (Table 2). Overall, the most commonlyused form of inpatient pharmacological prophylaxiswas enoxaparin (26.8% of all patients), followed byUFH (13.5% of all patients). Mechanical prophylaxiswas received by 12.2% of all patients. Mean 6 SDVTE prophylaxis duration during hospitalization was5.0 6 4.7 days, varying from 4.2 6 4.0 days inpatients with cancer to 6.2 6 5.5 days in patientswith infectious disease.In the 35 days after discharge, 8.8% of patients

received anticoagulation therapy, most commonlywarfarin (7.7%). The rate of outpatient prophylaxiswas highest in patients hospitalized for CHF (18.5%)compared with other medical conditions (�7%).

Time Course of VTE Risk and Hazard Function

Overall, there were 366 symptomatic VTE events, rep-resenting a VTE rate of 3.3%. These events comprised241 DVT-only events, 98 PE-only events, and 27events with evidence of both DVT and PE. In total,43.4% of events occurred during hospitalization (Fig-ure 2). The VTE rate was 5.7% in patients with can-cer (30.5% of events occurring in hospital), 4.3%with infectious disease (61.9% in hospital), 3.1%with CHF (54.7% in hospital), and 2.1% with severelung disease/COPD (42.6% in hospital). The highestnumber of VTE events, 97 events (62 DVT only, 26PE only, and 9 events both DVT and PE), occurred inthe first 9 days after the hospital admission date, ofwhich 87.6% were during hospitalization. Duringdays 10-19, there were 82 VTE events (50 DVT only,24 PE only, and 8 both DVT and PE), 70.7% ofwhich occurred in the hospital. Over the following10-day periods, VTE incidence gradually declined(Figure 2) and fluctuated at a background level of 4-8events during each 10-day interval from 120 to 180days.The cumulative probability of VTE among all

patients was 0.035 (Figure 3A). Half of the VTE riskhad accumulated by day 23, and 75% had accumulatedby day 71. By day 30, the proportion of cumulative riskwas 52.6% overall, and ranged from 41.9% with can-cer to 72.9% with infectious disease (Figure 3).The VTE hazard peaked at approximately 1.05 VTE

events per 1000 person-days on day 8 after the

FIG. 1. Flow diagram of patient inclusion. †Patients were continuously enrolled for at least 180 days after the hospital admission date, except if a patient became

disenrolled because of inpatient death from any cause; such patients were exempted from the continuous enrollment criterion as long as they had been

continuously enrolled before inpatient death. ‡The study sought an unbiased mix of patients with either a first at-risk hospitalization or a history of prior at-risk

hospitalizations; among patients with multiple at-risk hospitalizations during the patient selection period, the at-risk hospitalization on which the study focused

was randomly selected from among each of the candidate at-risk hospitalizations. Abbreviations: HDD, Hospital Drug Database; VTE, venous thromboembolism.

Duration of VTE Risk in Medically Ill Patients | Amin et al.


hospital admission date overall (Figure 4A). The cu-mulative hazard at the peak day was 18.2% of thetotal VTE hazard over the 180-day evaluation period.The hazard peak ranged from day 7 in patients with

severe lung disease/COPD to day 12 in patients withinfectious disease (Figure 4B). The cumulative hazardat the peak day was 39.7% for patients with infec-tious disease, 29.2% for patients with CHF, andapproximately 19% for cancer or severe lung disease/COPD. After the peak risk day, the VTE hazard func-tion decreased until the curve reached an inflectionpoint, at day 28, when the cumulative risk was51.8% (Figure 4A). After the inflection point, theVTE hazard increased to 0.3 VTE events per 1000person-days at approximately day 40-45 and thendecreased to <0.2 events per 1000 person-days. Thetiming of the inflection varied by approximately 1week across the medical illnesses (ranging from day25 for severe lung disease/COPD to day 33 for CHF),with the cumulative risk at the inflection point rang-ing from 41.9% with cancer to 72.9% with infectiousdisease.

DISCUSSIONThe results from this large, real-world study providenew insights into the duration of risk of symptomaticVTE in medical patients and demonstrate that thenumber of VTE events was highest during days 0-19,with the peak of VTE hazard at day 8. Half of thetotal 180-day cumulative risk had been incurred byday 23 after hospital admission, and the period ofgreatest increased risk extended up to at least 30days. Importantly, more than half of VTE eventsoccurred after discharge (56.6%). A particularly highproportion of VTE events (69.5%) had occurred afterdischarge in patients with cancer. Although it wasassumed that most VTE events that could be reason-ably attributed to an at-risk hospitalization wouldoccur within 90 days as shown previously,4,11 the180-day evaluation period was used to examinewhether there was a prolonged period of continuallydiminished VTE risk from 90 to 180 days. Thus,

TABLE 1. Summary of Patient Demoghraphics andCharacteristics

Characteristic Medical Patients (N ¼ 11,139)

GenderMen 5389 (48.4)Women 5750 (51.6)

Reason for hospitalizationCancer 2243 (20.1)CHF 1705 (15.3)Severe lung disease/COPD 5736 (51.5)Severe infectious disease 1455 (13.1)

Age group, years18–34 230 (2.1)35–44 442 (4.0)45–54 1188 (10.7)55–64 2644 (23.7)65–74 2657 (23.9)75–84 2969 (26.7)�85 years 1009 (9.1)

Median age 6 SD, years 67.6 6 13.9Primary payer*

Medicare 6819 (61.2)Commercial 4320 (38.8)

Geographical areaNortheast 122 (1.1)North Central 2649 (23.8)South 8258 (74.1)West 110 (1.0)

Urban location 9743 (87.5)Teaching hospital 1345 (12.1)Licensed bed size

1–199 1621 (14.6)200–299 2869 (25.8)300–499 4005 (36.0)�500 2644 (23.7)

NOTE: All values are presented as no. (%) unless indicated otherwise. 3Abbreviations: CHF, congestiveheart failure; COPD, chronic obstructive pulmonary disease; SD, standard deviation. * This study used onlythe Commercial and Medicare supplemental databases.

TABLE 2. DVT Prophylaxis Type During Hospitalization and in the Outpatient Setting 35 Days After Discharge

n (%)

Infectious Disease

(n ¼ 1455)

CHF

(n ¼ 1705)

Severe Lung

Disease/COPD (n ¼ 5736)

Cancer

(n ¼ 2243)

Any Medical

(N ¼ 11,139)

Pharmacological prophylaxis during hospitalization* 599 (41.2) 1093 (64.1) 2820 (49.2) 688 (30.7) 5200 (46.7)Enoxaparin 362 (24.9) 466 (27.3) 1877 (32.7) 282 (12.6) 2987 (26.8)UFH 191 (13.1) 400 (23.5) 527 (9.2) 383 (17.1) 1501 (13.5)Warfarin† 135 (9.3) 498 (29.2) 622 (10.8) 95 (4.2) 1350 (12.1)Dalteparin 16 (1.1) 21 (1.2) 109 (1.9) 16 (0.7) 162 (1.5)Fondaparinux 5 (0.3) 4 (0.2) 22 (0.4) 2 (0.1) 33 (0.3)

Mechanical prophylaxis in hospital 148 (10.2) 65 (3.8) 343 (6.0) 803 (35.8) 1359 (12.2)Anticoagulation within 35 days after discharge 104 (7.1) 315 (18.5) 397 (6.9) 166 (7.4) 982 (8.8)

Enoxaparin 15 (1.0) 14 (0.8) 32 (0.6) 25 (1.1) 86 (0.8)UFH 17 (1.2) 10 (0.6) 23 (0.4) 35 (1.6) 85 (0.8)Warfarin 79 (5.4) 302 (17.7) 357 (6.2) 116 (5.2) 854 (7.7)Dalteparin 0 0 2 (<0.1) 1 (<0.1) 3 (<0.1)Fondaparinux 1 (0.1) 0 0 2 (0.1) 3 (<0.1)

Antiplatelet therapy within 35 days after discharge 72 (4.9) 217 (12.7) 351 (6.1) 53 (2.4) 693 (6.2)

Abbreviations: CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; DVT, deep-vein thrombosis; UFH, unfractionated heparin. *With or without another prophylactic agent. †Patients with warfarin and noevidence of venous thromboembolism, with or without another prophylactic agent.



events occurring within the later portions of the evalu-ation period may or may not have been attributableto the index hospitalization, potentially reflecting a‘‘background’’ rate of VTE as noted above. Althoughthese events are included in our estimate of the180-day cumulative risk of VTE, interpretation of thestudy results excluding such events is possible byexamining the cumulative risk that had been incurredat each time point during the evaluation period.Few other studies have assessed the duration of VTE

risk in hospitalized medical patients. In a study bySpyropoulos et al,4 the median time to a DVT and/orPE event was 74 days, ranging from 62 days in severeinfectious disease to 126 days in CHF. In anotherobservational study that included patients who hadrecently been hospitalized but had not undergone sur-gery, 66.9% of patients who experienced DVT and/orPE events were diagnosed with DVT and/or PE withinthe first month after hospital discharge; 19.9%between months 1 and 2, and 13.2% between months2 and 3.12

Fewer than half of the patients in the present studyreceived thromboprophylaxis, which is consistent withother studies demonstrating the low prophylaxis ratesin medical inpatients.9,13–15 In a recently published USstudy of discharge records that included 22,455 medi-cal inpatients, prophylaxis rates were 59.4% inpatients with CHF, 52.3% with cancer, 45.8% withsevere lung disease/COPD, and 40.4% with infectiousdisease.14 Fewer than 10% of patients in the presentstudy received prophylaxis after discharge, a resultthat is consistent with other studies.4,9

The effect of extended prophylaxis in acutely illmedical patients with the LMWH enoxaparin beyond6-14 days has been investigated in the EXCLAIMstudy.16 This trial included approximately 5800acutely ill medical patients at significant risk of devel-oping VTE due to a recent reduction in mobility.Patients in the extended prophylaxis group had alower risk of VTE (2.5% vs 4% for placebo; absoluterisk reduction 1.5% [95.8% confidence interval

�2.54% to �0.52%]), but had increased major bleed-ing events (0.8% vs 0.3% for placebo; absolute riskdifference favoring placebo, 0.51% [95% confidenceinterval, 0.12% to 0.89%]). The patient populationswith most benefit from an additional 28 days prophy-laxis with enoxaparin, in addition to the usual short-term prophylaxis of 10 days, were patients with re-stricted mobility (level 1; total bed rest/sedentary), el-derly patients (age >75 years), and women. A limita-tion of the EXCLAIM trial is that estimates ofefficacy and safety are difficult to interpret: after aninterim analysis of adjudicated efficacy and safety out-comes, amendments were made to the original studyprotocol by changing eligibility criteria for patientswith level 2 immobility (level 1 with bathroomprivileges).16

The optimal duration of prophylaxis for medicalpatients has not been determined; prophylaxis is gen-erally administered to at-risk medical patients for theduration of hospitalization. In the current study, meanlength of stay was 5.3 6 5.3 days overall. As hospitalstays shorten, many medical patients who are pre-scribed inpatient prophylaxis alone are unlikely toreceive the standard 6-14 days of prophylaxis shownto be effective in clinical trials.6–8 Furthermore, theextended period of VTE risk in the present study andthe finding that 56.6% of events occurred after dis-charge also suggest that current practices for inpatientprophylaxis alone may need to be evaluated.This study reports real-world data from a large,

well-defined population and obtains the incidence ofsymptomatic VTE events. Even though certain demo-graphic data deviate from the national average—forexample, 74.1% of patients were treated in the SouthCensus region, whereas this region is served by 37.6%of US hospitals17; 87.5% of hospitals had an urbanlocation (compared with 60.1% of US hospitals18),and 85.4% of hospitals had a licensed bed size of atleast 200 beds (compared with 28.2% of US hospitals,with the average US hospital having fewer than 100beds19)—these data may be beneficial in guiding pol-icy and health care strategies for gaining understand-ing of the duration of risk for VTE.Limitations of the study include characterization of

the VTE risk period through examination of the cumu-lative risk and hazard of VTE across time, as the actualVTE risk period cannot be determined with exact pre-cision. We used ICD-9-CM diagnosis coding to identifyVTE. Since many cases of PE are asymptomatic anddetected at autopsy,20 our approach may have missedsuch cases, as they would not have been recordedwithin the database. Furthermore, validation studiessuggest that suboptimal specificity exists for ICD-9-CM diagnosis codes used to identify VTE.21 In anattempt to improve the specificity of our VTE identifi-cation algorithm, we required that post-discharge VTEwas recorded either during an emergency room or sub-sequent inpatient admission (which would be indicative

FIG. 2. Distribution of VTE risk events occurring during hospitalization

versus after discharge is displayed on the left-hand y-axis. The absolute and

cumulative numbers of VTE events incurred during each 10-day period are

displayed on the right-hand y-axis. Abbreviation: VTE, venous

thromboembolism.



of acute care for VTE) or on an outpatient claim withsubsequent evidence of treatment for VTE. The truesensitivity and specificity of the VTE identificationalgorithms used for this study remain unknown,however, so the study findings should be interpreted inlight of this limitation. The databases used for the anal-ysis may not be representative of the US population asa whole; for example, this study used claims data fromcommercial and Medicare supplemental databases,which do not include Medicaid patients. Another limi-tation was that outpatient mechanical prophylaxis, suchas graded compression stockings, was not captured dueto over-the-counter availability. In addition, appropri-

ateness of prophylaxis was not determined in this study,

because these data could not be obtained from the

claims database used. Further studies are warranted to

obtain information on the incidence of VTE after hospi-talization for medical illness in patients who receivedappropriate prophylaxis during hospitalization.Finally, all dosages of a pharmacological agent were

considered prophylactic only if a VTE event did notoccur, with the exception of warfarin; any dose ofwarfarin was considered for prophylaxis, regardless ofa VTE diagnosis. Warfarin may be used for purposesother than VTE prophylaxis (eg, prophylaxis for athromboembolic cerebrovascular accident). The data

FIG. 3. Cumulative probability of VTE by number of days after hospital admission for the total cohort of medically ill patients (A) and for each medical illness (B).

The left-hand y-axis shows the percentage of the total 180-day cumulative probability of VTE; the right-hand y-axis shows the absolute cumulative probability of

VTE. The x-axis shows the number of days after the hospital admission date. Abbreviations: CHF, congestive heart disease; COPD, chronic obstructive

pulmonary disease; VTE, venous thromboembolism.



source does not allow for identifying the exact reasonfor anticoagulation therapy with warfarin. Nonethe-less, warfarin therapy will confer a decreased risk ofVTE regardless of its purpose.Results from this large cohort of medical patients

indicate that symptomatic VTE risk is highest withinthe first 19 days after hospital admission (a period thatmay encompass both the duration of hospitalization aswell as the period after discharge) with a considerablerisk of VTE extending into the period after discharge.Receiving appropriate prophylaxis in-hospital remainsof great importance to prevent inpatient and likelypost-discharge VTE in patients with acute medical ill-

ness. In addition, given the time course of VTE events,with VTE incidence peaking at 8 days but withincreased risk extending to 30 days, and the number ofout-of-hospital VTE events incurred, the results of thisstudy suggest that future research is warranted to inves-tigate the risks and benefits of improving thrombopro-phylaxis practices in the period after hospitalization.

AcknowledgementsFunding Source: sanofi-aventis U.S. provided funding to Thomson Reu-ters to perform this study. The authors received editorial/writing supportfrom Tessa Hartog of Excerpta Medica in the preparation of the manu-script funded by sanofi-aventis U.S.

FIG. 4. Hazard function of VTE by number of days after hospital admission for all medically ill patients (A) and for each medical illness (B). Abbreviations: CHF,

congestive heart disease; COPD, chronic obstructive pulmonary disease; VTE, venous thromboembolism.



Disclosure: Alpesh Amin has received research honorarium and is on thespeakers bureau for sanofi-aventis U.S. Otsuka Pharmaceutical, andBoehringer-Ingelheim. Helen Varker, Nicole Princic, and Stephen John-ston are employees at Thomson Reuters, which received funding fromsanofi-aventis U.S. Jay Lin is an employee of Novosys Health, whichreceived funding from sanofi-aventis U.S. Stephen Thompson is an em-ployee of sanofi-aventis U.S.

References1. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous

thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6Suppl):381S–453S.

2. Heit JA, Silverstein MD, Mohr DN, Petterson TM, O’Fallon WM,Melton LJ 3rd. Risk factors for deep vein thrombosis and pulmonaryembolism: a population-based case-control study. Arch Intern Med.2000;160:809–815.

3. Heit JA. The epidemiology of venous thromboembolism in the com-munity. Arterioscler Thromb Vasc Biol. 2008;28:370–372.

4. Spyropoulos AC, Hussein M, Lin J, Battleman D. Rates of venousthromboembolism occurrence in medical patients among the insuredpopulation. Thromb Haemost. 2009;102:951–957.

5. Heit JA, O’Fallon WM, Petterson TM, et al. Relative impact of riskfactors for deep vein thrombosis and pulmonary embolism: a popula-tion-based study. Arch Intern Med. 2002;162:1245–1248.

6. Samama MM, Cohen AT, Darmon JY, et al. A comparison of enoxa-parin with placebo for the prevention of venous thromboembolism inacutely ill medical patients. Prophylaxis in Medical Patients withEnoxaparin Study Group. N Engl J Med. 1999;341:793–800.

7. Leizorovicz A, Cohen AT, Turpie AG, et al. Randomized, placebo-con-trolled trial of dalteparin for the prevention of venous thromboembo-lism in acutely ill medical patients. Circulation. 2004;110:874–879.

8. Cohen AT, Davidson BL, Gallus AS, et al. Efficacy and safety of fon-daparinux for the prevention of venous thromboembolism in olderacute medical patients: randomised placebo controlled trial. BMJ.2006;332:325–329.

9. Amin A, Lin J, Ryan A. Lack of thromboprophylaxis across the carecontinuum in US medical patients. Hosp Pract (Minneap). 2010;38:17–25.

10. HCUP NIS Related Reports. Healthcare Cost and Utilization Project(HCUP), September 2008. Available at: www.hcup-us.ahrq.gov/db/nation/nis/nisrelatedreports.jsp. Accessed June 2011.

11. Spyropoulos AC, Lin J. Direct medical costs of venous thromboembo-lism and subsequent hospital readmission rates: an administrativeclaims analysis from 30 managed care organizations. J Manag CarePharm. 2007;13:475–486.

12. Spencer FA, Lessard D, Emery C, Reed G, Goldberg RJ. Venousthromboembolism in the outpatient setting. Arch Intern Med. 2007;167:1471–1475.

13. Amin AN, Stemkowski S, Lin J, Yang G. Inpatient thromboprophy-laxis use in U.S. hospitals: adherence to the seventh American Collegeof Chest Physician’s recommendations for at-risk medical and surgi-cal patients. J Hosp Med. 2009;4:E15–E21.

14. Amin A, Spyropoulos AC, Dobesh P, et al. Are hospitals deliveringappropriate VTE prevention? The venous thromboembolism study toassess the rate of thromboprophylaxis (VTE start). J Thromb Throm-bolysis. 2010;29:326–339.

15. Bergmann JF, Cohen AT, Tapson VF, et al. Venous thromboemb-olism risk and prophylaxis in hospitalised medically ill patients.The ENDORSE Global Survey. Thromb Haemost. 2010;103:736–748.

16. Hull RD, Schellong SM, Tapson VF, et al. Extended-duration venousthromboembolism prophylaxis in acutely ill medical patients withrecently reduced mobility: a randomized trial. Ann Intern Med. 2010;153:8–18.

17. American Society for Healthcare Engineering of the American HospitalAssociation. Overview of the Hospital Market, 2009. Available from:www.ashe.org/e2c/pdfs/energy/heg_ch2_background.pdf. AccessedJune 2011.

18. American Hospital Association. Fast Facts on US Hospitals, 2009.Available at: http://www.aha.org/aha/resource-center/Statistics-and-Studies/fast-facts.html. Accessed June 2011.

19. American Hospital Association. AHA Annual Survey of HospitalsDatabase, 2009. Available from: http://www.ahadata.com/ahada-ta_app/index.jsp. Accessed June 2011.

20. White RH. The epidemiology of venous thromboembolism. Circula-tion. 2003;107(23 suppl 1):I4–I8.

21. White RH, Sadeghi B, Tancredi DJ, et al. How valid is the ICD-9-CMbased AHRQ patient safety indicator for postoperative venousthromboembolism? Med Care. 2009;47:1237–1243.



Documents

Duration of Venous Thromboembolism Risk Across a Continuum ... · ORIGINAL RESEARCH Duration of Venous Thromboembolism Risk Across a Continuum in Medically Ill Hospitalized Patients